def test_space_ends_word(self):
# "cit ʔa" being parsed as
# 440. iˑ vowel
# 441. c consonant
# 442. i missing +
# 443. t ʔ missing *
# 444. a vowel
f = FileReader()
f.data['consonants'] = f.parse_inventory("c, t(t, tⁿ)", 'consonant')
f.data['vowels'] = f.parse_inventory("iˑ, a", 'vowel')
# add known missings
f.known_missings = [
MissingToken("i", known_missing=True),
MissingToken("ʔ", known_missing=True),
]
transcript = f.standardise("iˑcit ʔa")
parsed = f.parse_transcript(transcript)
assert parsed[0] == Token("iˑ")
assert parsed[1] == Token("c")
assert parsed[2] == MissingToken("i", known_missing=True)
assert parsed[3] == Token("t(t, tⁿ)")
assert parsed[4] == Token(" ")
assert parsed[5] == MissingToken("ʔ", known_missing=True)
assert parsed[6] == Token("a")
def test_missing_token(self):
string = "ɔlaɣ"
parsed = self.reader.parse_transcript(string)
assert len(parsed) == len(string)
assert parsed[0] == MissingToken("ɔ")
assert parsed[1] == Token("l")
assert parsed[2] == Token("a(a, ɑ, ə, æ)")
assert parsed[3] == MissingToken("ɣ")
def test_slippage_and_duplication(self):
string = "ɔlalɣ"
parsed = self.reader.parse_transcript(string)
assert len(parsed) == len(string)
assert parsed[0] == MissingToken("ɔ")
assert parsed[1] == Token("l")
assert parsed[2] == Token("a(a, ɑ, ə, æ)")
assert parsed[3] == Token("l")
assert parsed[4] == MissingToken("ɣ")
def test_basaa_ignored_superscript_n(self):
# gáː ⁿbɛ̀βí being parsed as
#
# 9. h consonant
# 10. a(a, á, à, ǎ, â) vowel
# 11. ŋ(ŋ, ŋ́, ŋ̀) consonant
# 12. g missing *
# 13. aː(aː, áː, àː, ǎː, âː) vowel
# 14. ⁿ missing *
# 15. b missing *
# 16. ɛ(ɛ, ɛ́, ɛ̀, ɛ̌, ɛ̂) vowel
# 17. β consonant
# 18. i(i, í, ì, ǐ, î) vowel
#
# i.e. 14 should be combined with 15 = ⁿb
f = FileReader()
f.data['consonants'] = f.parse_inventory("gʷ, ⁿb, ⁿg, β", 'consonant')
f.data['vowels'] = f.parse_inventory(
"""
a(a, á, à, ǎ, â), aː(aː, áː, àː, ǎː, âː),
e(e, é, è, ě, ê), ɛ(ɛ, ɛ́, ɛ̀, ɛ̌, ɛ̂),
i(i, í, ì, ǐ, î),
""", 'vowels')
transcript = f.standardise('gáː ⁿbɛ̀βí')
parsed = f.parse_transcript(transcript)
assert parsed[0] == MissingToken("g") # known missing
assert parsed[1] == Token("aː(aː, áː, àː, ǎː, âː)")
assert parsed[2] == Token(" ") # was incorrect -- should be SPACE.
assert parsed[3] == Token("ⁿb") # was incorrect
assert parsed[4] == Token("ɛ(ɛ, ɛ́, ɛ̀, ɛ̌, ɛ̂)")
assert parsed[5] == Token("β")
assert parsed[6] == Token("i(i, í, ì, ǐ, î)")
def test_combining_in_others(self):
# Setswana's 'bó̝kɔ̝̀ːnì' was being parsed as:
#
# 8. b consonant
# 9. o̝(o̝, ò̝, ó̝, ô̝, ǒ̝) vowel
# 10. k consonant
# 11. ɔ̝̀ː missing *
# 12. n consonant
# 13. i(i, ì, í, î, ǐ, ì̞, í̞) vowel
#
# i.e. in token 11 the combining character of double triangle "ː" is
# merged to the character 'ɔ̝̀'. 'ɔ̝̀' is IN the inventory, but 'ɔ̝̀ː' is NOT
# so this gets flagged as an error. "ː" is in other symbols and is
# currently not being recognized as such
f = FileReader()
f.data['consonants'] = f.parse_inventory("b, k, n", 'consonant')
f.data['vowels'] = f.parse_inventory(
"o̝(o̝, ò̝, ó̝, ô̝, ǒ̝), i(i, ì, í, î, ǐ, ì̞, í̞)", 'vowel')
f.known_missings.update(f.parse_list(['/ɔ̝̀/']))
f.other_symbols.update(f.parse_inventory('ː', 'other'))
# Other: ː
transcript = 'bó̝kɔ̝̀ːnì'
transcript = f.standardise(transcript)
parsed = f.parse_transcript(transcript)
assert parsed[0] == Token("b"), parsed
assert parsed[1] == Token("o̝(o̝, ò̝, ó̝, ô̝, ǒ̝)"), parsed
assert parsed[2] == Token("k"), parsed
assert parsed[3] == MissingToken("ɔ̝̀"), parsed
assert parsed[4] == Token("ː"), parsed
assert parsed[5] == Token("n"), parsed
assert parsed[6] == Token("i(i, ì, í, î, ǐ, ì̞, í̞)"), parsed
def test_danish_overextension(self):
# being parsed as ... MissingToken("də") not MissingToken("d"),
# Token("ə")
transcript = 'b̥lɛːsdə'
f = FileReader()
f.data['consonants'] = f.parse_inventory(
"b̥(b̥, b̥ʰ), d̥(d̥, d̥s), s, l(l, l̩)", 'consonant')
f.data['vowels'] = f.parse_inventory("e(e, eː), ɛ(ɛ, ɛː), a, ɑ, ə",
'vowel')
transcript = f.standardise(transcript)
parsed = f.parse_transcript(transcript)
assert parsed[0] == MissingToken("b̥(b̥, b̥ʰ)")
assert parsed[1] == Token("l(l, l̩)")
assert parsed[2] == Token("ɛ(ɛ, ɛː)")
assert parsed[3] == Token("s")
assert parsed[4] == MissingToken("d")
assert parsed[5] == Token("ə")
def test_get_missing(self):
# if missing char is in the known_missings, then it returns a MissingToken
# with known_missing set to True
x = self.reader.get_missing('x')
assert x == MissingToken('x')
assert x.known_missing == True
# if missing char is in the default_tokens, then it returns a Token
# with phoneme_type="default"
dot = self.reader.get_missing(".")
assert dot == Token('.')
assert dot.is_missing == False
assert dot.phoneme_type == 'default'
# otherwise it's just Missing
nine = self.reader.get_missing('9')
assert nine == MissingToken('9')
assert nine.known_missing == False
def test_identified_missing_characters(self):
expected_missings = [
Token("oː"),
Token("ɣ"),
Token("ɔ"),
Token("ç"),
MissingToken('ɂ')
]
for m in self.reader.errors:
assert m in expected_missings, '%s is not in expected_missings' % m
def test_one(self):
# parsed as
#130. j(j, ɲ) consonant
#131. o(o, ó, ò, ô, ŏ̀) vowel
#132. ↗ other
#133. óʔẁ̰ missing *
#134. punctuation
transcript = self.f.standardise("jó↗óʔẁ̰")
parsed = self.f.parse_transcript(transcript)
assert parsed[0] == Token("j(j, ɲ)")
assert parsed[1] == Token("o(o, ó, ò, ô, ŏ̀)")
assert parsed[2] == MissingToken("↗")
assert parsed[3] == Token("o(o, ó, ò, ô, ŏ̀)")
assert parsed[4] == Token("ʔw̰(ʔw̰, ʔŋ̰ʷ, ʔẁ̰)")
def test_maximal_error(self):
# should identify the missing token as "o:" not ":"
transcript = 'oːlal'
f = FileReader()
f.data['consonants'] = f.parse_inventory("l, ɭ, ʎ, r(r, ɾ, ɹ)",
'consonant')
f.data['vowels'] = f.parse_inventory(
"a(a, ɑ, ə, æ), o(o, ɒ), u(u, ʊ), uː", 'vowel')
transcript = f.standardise(transcript)
parsed = f.parse_transcript(transcript)
assert parsed[0] == MissingToken("oː")
assert parsed[1] == Token("l")
assert parsed[2] == Token("a(a, ɑ, ə, æ)")
assert parsed[3] == Token("l")
def test_known_missing(self):
assert MissingToken("X", known_missing=True).known_missing
assert not MissingToken("X", known_missing=False).known_missing
def test_repr_known_missing(self):
assert repr(MissingToken(
"X", known_missing=True)) == "<KnownMissingToken: X>"
def test_repr(self):
assert repr(MissingToken("X")) == "<MissingToken: X>"
def test_phoneme_type(self):
assert MissingToken("X").phoneme_type == 'missing'
def test_is_missing(self):
assert MissingToken("X").is_missing
def test_missing_token(self):
MissingToken()
def test_other_symbols_not_in_missing(self):
assert MissingToken("ɂ") not in self.reader.errors
def test_not_greedy(self):
errors = self.reader.parse_list(["/x/ something/or other"])
assert errors == [MissingToken("x", known_missing=True)]
assert errors[0].is_missing == True
assert errors[0].known_missing == True
def test_simple_error(self):
errors = self.reader.parse_list(["/x/ something"])
assert errors == [MissingToken("x", known_missing=True)]
assert errors[0].is_missing == True
assert errors[0].known_missing == True
